The focus this year was on two assays conducted in the field: direct skin feeding assays (application of unfed mosquitoes to a human volunteer) and experimental hut assays (collection of mosquitoes from the home of a volunteer). These assays were conducted under NIAID protocol 15-I-0044, a Phase 1 study on the safety and immunogenicity of Pfs230 and Pfs25 TBV candidates against Plasmodium falciparum. In total, 1,844 direct skin feeding assays were performed with an infection rate of 1.6%%. To our knowledge, this number of feeding assays represents the largest scale implementation of DSFs worldwide in any malaria trial, and requires vast investment in mosquito production, maintenance, and dissection. Mosquito specimens were preserved from each feeding assay to investigate parasite development and develop molecular assays that assess transmission-blocking activity. Current molecular markers being tested against field samples are the parasite and gametocyte markers 18, Pfs25, and Nek3, as well as the ookinete specific genes SOAP and WARP. All of these markers were successfully detected in field-collected specimens, providing reproducible and robust parasite and ookinete detection. Currently, direct skin feeding assays are conducted on all trial volunteers and one of the goals of this project is to confirm parasite gene signatures that predict infectivity to mosquitoes, allowing feeds to be conducted only on those individuals who are truly infectious at a given timepoint. As part of this work, peripheral blood samples were collected from all individuals at every direct skin feeding to test against a panel of gametocyte marker genes: up in infectious, down in infectious, and constant housekeeping. The panel of genes used were shown to be a very robust predictor of infectious feeds but less robust in non-infectious feeds and feeds originating from individuals with low gametocyte levels. Overall, none of these gene signatures were found to be a better predictor than gametocyte level from a blood smear. To date, over 400 experimental hut assays have been conducted on volunteers enrolled in a number of different clinical studies in Mali: NIAID protocols 11-I-N143, 14-I-N159 (closed studies), 13-I-N109, and 15-I-0044. Samples from some of these studies have been analyzed for feeding fidelity i.e. genetic comparison of the identity of the mosquito bloodmeal source with the genetic profile of the individual sleeping in the hut. Feeding fidelity is crucial to determine the validity of using experimental huts as a TBV efficacy surrogate. In 2017, fidelity of the 2015 transmission season mosquitoes was completed, with a rate of 65% in mosquitoes collected from huts in Bancoumana. Parasite detection was performed in all experimental huts from multiple seasons with rates of parasite detection of 48% in Bancoumana in the 2014 season, 55% in Doneguebougou in the 2014 season, and 69% in Bancoumana in the 2015 season. In conclusion, it was determined that while experimental huts are very feasible to perform and well accepted in the Malian communities of Bancoumana and Doneguebougou, they do not provide a suitable number of endpoints to use in TBV assessment going forward.